Data Manipulation Exercises

Load Packages

Load the tidyverse package and the ggplot2 package.

library(tidyverse)

## ── Attaching packages ─────────────────────────── tidyverse 1.3.0 ──

## ✓ ggplot2 3.3.2     ✓ purrr   0.3.4
## ✓ tibble  3.0.1     ✓ dplyr   1.0.0
## ✓ tidyr   1.1.0     ✓ stringr 1.4.0
## ✓ readr   1.3.1     ✓ forcats 0.5.0

## Warning: package 'ggplot2' was built under R version 3.6.2

## Warning: package 'tibble' was built under R version 3.6.2

## Warning: package 'tidyr' was built under R version 3.6.2

## Warning: package 'purrr' was built under R version 3.6.2

## Warning: package 'dplyr' was built under R version 3.6.2

## ── Conflicts ────────────────────────────── tidyverse_conflicts() ──
## x dplyr::filter() masks stats::filter()
## x dplyr::lag()    masks stats::lag()

library(ggplot2)

Import data: chds6162_data

Import the chds6162_data to a dataframe called data.

data <- read_csv("data/chds6162_data.csv")

## Parsed with column specification:
## cols(
##   .default = col_double(),
##   drace = col_character()
## )

## See spec(...) for full column specifications.

select

Use select to show just the marital variable.

data %>% 
    select(marital)

select for multiple variables:

Use select to show marital and ed variables.

data %>% 
    select(marital,ed)

Use select for a range of columns.

select all the variables from wt to the end.

# option one:

data %>% 
    select(wt:number)

# option 2:

data %>% 
    select(wt:last_col())

Drop the race variable.

data %>% 
    select(-race)

Drop the variables that belonged to the father from drace to dwt

data %>% 
    select(-(drace:dwt))

mutate

art by @allison_horst

Create a new variable with a specific value

Create a new variable called country and fill that variable with “US”.

data %>% 
    mutate(country = "US") 

Create a new variable based on other variables

Create a new variable called gestation_weeks and calculate gestation length in weeks rather than days (try rounding this number to only 2 decimals). Remember that the gestation variable is a measure of the gestation length in days. Then select both variables.

data %>% 
    mutate(gestation_weeks = round(gestation / 7,2)) %>% 
    select(gestation, gestation_weeks)

Change an existing variable

Convert the dwt variable to kilos by dividing by 2.205 (it’s in pounds now). Then, use select to show only the dwt variable.

data %>% 
    mutate(dwt = dwt / 2.205) %>% 
    select(dwt)

case_when

art by @allison_horst

Using mutate and case_when, let’s create a variable called age_group. In this variable we will use the mom’s age variable and recode it so all moms whose age are now “20s”, “30s”, and so on. Then select the id and the new variable

*tip= Wanna know the range of your sample age? Use range(dataframe$variable,na.rm = TRUE)

** Use %in% (you just learned it 😉)

range(data$age,na.rm = TRUE)

## [1] 15 45

data %>% 
  mutate(age_group = case_when(
    age %in% 10:19 ~ "10s",
    age %in% 20:29 ~ "20s",
    age %in% 30:39 ~ "30s",
    age %in% 40:49 ~ "40s",
    age %in% 50:59 ~ "50s"
  ))

filter

art by @allison_horst

We use filter to choose a subset of cases.

Use filter to keep only respondents who are divorced (marital = 3).

data %>% 
    filter(marital == 3) %>% 
    select(marital)

Use filter to keep only respondents who are not divorced.

data %>% 
    filter(marital != 3) %>% 
    select(marital)

Use %in% within the filter function to keep only those whose mothers’ race was reported as white. In this dataset, values from 0 to 5 represent “white” under the racevariable.

data %>% 
    filter(race %in% c(0:5))

Create a chain that keeps only those are college grads (value of 5). Then, filter to keep only those who are not married (value of 5). Finally, use select to show only the ed and marital variables.

data %>% 
    filter(ed == 5) %>% 
    filter(marital != 5) %>% 
    select(ed, marital)

summarize or summarise

Get the mean, median, min, and max weight for the fathers dwt.

data %>% 
    summarize(mean_dad_weight = mean(dwt,na.rm = TRUE),
              median_dad_weight = median(dwt, na.rm = TRUE),
              min_dad_weight = min(dwt, na.rm = TRUE),
              max_dad_weight = max(dwt, na.rm = TRUE))

group_by

art by @allison_horst

Calculate the mean weight for fathers based on their education levels (dedvariable has some NAs, don’t forget to remove them!) with na.rm=)

data %>% 
    group_by(ded) %>% 
    drop_na(ded) %>%
    summarize(dad_weight = mean(dwt,
                                na.rm = TRUE))

## `summarise()` ungrouping output (override with `.groups` argument)

across

art by @allison_horst

We messed up and we found out that out collaborators need the mean of all the variables for dad by their race category (dads’).

Use across to get them all! * Don’t forget to remove NAs otherwise it will yell at you!

data %>% 
  drop_na(drace) %>%
  group_by(drace) %>%
  summarize(across(dage:dwt,mean,na.rm = TRUE))

## `summarise()` ungrouping output (override with `.groups` argument)

Big exercise

A new data frame called mothers_below30 that:

1. Uses filter to only include those age younger than 30
2. Uses mutate to transform the gestation variable into weeks rather than days
3. Uses group_by to create mom education groups
4. Uses summarise to calculate a new variable called mean_gestation_w
5. Make sure to export it as a csv!

• Once you have that data frame just move of the variables to a different position using “relocate”

mothers_below30 <- data %>% 
    filter(age < 30) %>% 
    mutate(gestation_w = gestation/7) %>%
    group_by(ed) %>% 
    summarize(mean_gestation_w = mean(gestation_w,na.rm = TRUE))

## `summarise()` ungrouping output (override with `.groups` argument)

write_csv(mothers_below30,"exports/moms_under30.csv")

With Bonus:

mothers_below30 <- data %>% 
    filter(age < 30) %>% 
    mutate(gestation_w = gestation/7) %>%
    group_by(ed) %>% 
    summarize(mean_gestation_w = mean(gestation_w,na.rm = TRUE)) %>%
    relocate (mean_gestation_w, .after=ed)

## `summarise()` ungrouping output (override with `.groups` argument)

write_csv(mothers_below30,"exports/moms_under30.csv")